Mobile Internet Anomaly Traffic Detection Technology Research based on Improved Wavelet Neural Network

摘要

Mobile Internet anomaly traffic detection has a very important significance for ensuring the effective operation of the network and raising the robustness of service providing ability. Quantum particle swarm optimization algorithm is combined with the wavelet neural network, and the parameters of the neural network are optimized by quantum particle swarm optimization. We build a wavelet neural network model based on quantum particle swarm algorithm. Because quantum particle swarm optimization algorithm is easy to fall into local optimum, and it reduces the diversity of population and the global search ability at the same time, we put forward improved quantum particle swarm optimization algorithm based on adaptive local search scheme. Then the improved scheme is used to optimize parameters of wavelet neural network. The experiment results show that the proposed scheme has higher detection rate of abnormal state and lower misjudgment rate of normal state than tradition quantum particle swarm optimization algorithm.